In the mammalian genome, ~70% of cytosines within 5'-CpG-3' dinucleotides are methylated, an epigenetic modification correlating with transcriptional silencing. Twice in pluripotent early development, DNA methylation is globally erased. It is then re-established and maintained throughout somatic cell differentiation. Until recently, no global methylation changes were thought to occur in normal somatic cells. This long-held dogma was overturned by our discovery that erythroid differentiation is associated with DNA demethylation at nearly all genomic elements. Global demethylation is required for rapid erythroid gene induction. It represents a novel genome-wide epigenetic transformation essential for erythroid maturation, whose mechanism is as yet unknown. We hypothesize that disruption of global erythroid demethylation may underlie syndromes of erythropoietin-resistant anemias. Indeed, genome-wide aberrant DNA methylation was recently found in the refractory anemias classified as myelodysplastic syndromes (MDS), where demethylation therapy is successful. Our long-term goal is to elucidate the mechanism and function of global DNA demethylation in erythropoiesis and the consequences of its dysregulation. During replication, DNA methyltransferase 1 (Dnmt1) methylates nascent DNA, thereby maintaining DNA methylation across cell generations. Erythroid global demethylation is dependent on DNA replication, suggesting impaired Dnmt1 function. Demethylation further requires an accelerated intra-S phase DNA synthesis, a recently-discovered cell cycle behavior at the onset of erythroid differentiation whose mechanism is unknown. Demethylation in erythroblasts is not, however, associated with Dnmt1 loss, and cannot be prevented by Dnmt1 overexpression. Together, these observations suggest the hypothesis that erythroblasts contain mechanisms limiting the methylation capacity of Dnmt1. We propose to test this hypothesis with the following three aims: Aim 1: Determine whether 5mC in erythroid genomic DNA is subject to oxidation by the enzyme Tet2, resulting in replication-dependent, Dnmt1-resistant global demethylation Aim 2: Determine the role of the Cyclin-dependent kinase inhibitor p57KIP2 in S phase acceleration and in erythroid global DNA demethylation. Aim 3: Determine the underlying mechanism of the accelerated intra-S phase DNA synthesis at the onset of erythroid differentiation: is it caused by an increased number of firing replication origins, or by increased rate of replication fork progression. Ask whether this process is regulated by the cyclin-dependent kinase inhibitor p57KIP2. Funding of this work will elucidate how global methylation patterns are regulated, and may provide conceptually new avenues for therapy of erythropoietin-resistant anemias.